Download Spatially resolved measurement of femtosecond

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Diffraction topography wikipedia , lookup

Microscopy wikipedia , lookup

Rutherford backscattering spectrometry wikipedia , lookup

Super-resolution microscopy wikipedia , lookup

Vibrational analysis with scanning probe microscopy wikipedia , lookup

Photonic laser thruster wikipedia , lookup

Optical tweezers wikipedia , lookup

Optical aberration wikipedia , lookup

Ultraviolet–visible spectroscopy wikipedia , lookup

Confocal microscopy wikipedia , lookup

3D optical data storage wikipedia , lookup

Harold Hopkins (physicist) wikipedia , lookup

Silicon photonics wikipedia , lookup

Interferometry wikipedia , lookup

Nonimaging optics wikipedia , lookup

Ellipsometry wikipedia , lookup

Ultrafast laser spectroscopy wikipedia , lookup

Photon scanning microscopy wikipedia , lookup

Nonlinear optics wikipedia , lookup

Retroreflector wikipedia , lookup

Surface plasmon resonance microscopy wikipedia , lookup

Birefringence wikipedia , lookup

Diffraction grating wikipedia , lookup

Phase-contrast X-ray imaging wikipedia , lookup

Dispersion staining wikipedia , lookup

Fiber Bragg grating wikipedia , lookup

Anti-reflective coating wikipedia , lookup

Refractive index wikipedia , lookup

Transcript
July 15, 2012 / Vol. 37, No. 14 / OPTICS LETTERS
3003
Spatially resolved measurement of femtosecond
laser induced refractive
index changes in transparent materials
René Berlich,1,2 Jiyeon Choi,1,3 Clarisse Mazuir,4 Winston V. Schoenfeld,4
Stefan Nolte,2 and Martin Richardson1,*
1
Townes Laser Institute, College of Optics and Photonics, University of Central Florida, 4000 Central Florida Boulevard,
Orlando, Florida 32816, USA
2
Institute of Applied Physics, Friedrich Schiller University, Max-Wien-Platz 1, 07743 Jena, Germany
3
Now at the Department of Laser & Electron Beam Application, Korea Institute of Machinery & Materials,
104 Sinseongno, Yuseong-Gu, Daejeon, 305-343, Korea
4
CREOL, College of Optics and Photonics, University of Central Florida, 4000 Central Florida Boulevard,
Orlando, Florida 32816, USA
*Corresponding author: [email protected]
Received March 30, 2012; revised June 1, 2012; accepted June 1, 2012;
posted June 5, 2012 (Doc. ID 165857); published July 13, 2012
We present a practical method to determine femtosecond laser induced refractive index changes in transparent
materials. Based on an iterative Fourier transform algorithm, this technique spatially resolves the refractive index
of complex structures by combining the dimensions of the modified region with the corresponding phase change
extracted from far-field intensity measurements. This approach is used to characterize optical waveguides written
by a femtosecond laser in borosilicate glass. © 2012 Optical Society of America
OCIS codes: 100.5070, 140.3390, 290.3030.
Femtosecond laser direct writing (FLDW) has become a
common tool to fabricate three-dimensional photonic
devices in transparent materials [1,2]. Recent investigations show its potential for creating volumetric
diffractive optical elements (DOEs) [3–5] capable of multiplexing spatial and spectral information. However, for
efficient design of those elements, accurate characterization of the amount and the spatial distribution of the laser
induced refractive index change (Δn) is essential.
Furthermore, for each application, different writing parameters may need to be adapted and optimized so that a
practical and flexible refractive index measurement technique is required for the customized fabrication of DOEs.
It is common to deduce Δn from measurements of the
diffraction efficiency of a grating [6] or from the NA of a
written waveguide [7]. Both approaches are practical
since they are fast and do not require special analysis
equipment. However, they are limited in scope since
they are based on approximations, neither providing
data for complex structures nor any spatial information.
Other methods, which investigate, for instance, the reflectivity of a volume Bragg grating [8] or the mode
profile of a waveguide [9], yield better accuracy as well
as good spatial resolution, but are limited to special
structures. Refractive near-field profilometry [1] or selective chemical etching [10] techniques provide a high
sensitivity (Δn 10−5 relative and 10−4 absolute) and
a high spatial resolution (∼0.5 μm). However, they are
costly and require additional sample preparation. To
our knowledge, at the present time there is no practical,
fast method to spatially resolve femtosecond laser induced refractive index changes with high measurement
accuracy.
In this Letter, we present a novel solution to this need.
When a given FLDW structure is illuminated with a laser
beam of wavelength λ [Fig. 1(a)], the interaction can be
0146-9592/12/143003-03$15.00/0
described using a thin element approximation (TEA).
Within this approximation, diffraction effects during
the light propagation through the structure can be neglected. This approach can be applied because the femtosecond laser interaction induces only a small change in
the refractive index over spatial features with sizes large
compared to the wavelength λ [11]. By using TEA, the
value of the field following the structure uout x; y; z0 is obtained through multiplication of the incoming
field uin x; y; z0 with a transfer function tx; y. For a
nonabsorbent structure, this gives
uout x; y; z0 uin x; y; z0 · tx; y
uin x; y; z0 · expiΔϕx; y;
(1)
where Δφx; y is the induced phase change and is
related to the refractive index structure nx; y; z by
Fig. 1. (Color online) Experimental setup to measure the farfield intensity pattern caused by an induced refractive index
change distribution replicated in x direction.
© 2012 Optical Society of America
3004
OPTICS LETTERS / Vol. 37, No. 14 / July 15, 2012
Δϕx; y 2π
λ
Z
d
0
Δnx; y; zdz 2π
· Δneff x; y · d: (2)
λ
Here, Δneff x; y is an effective refractive index change
averaged in the propagation direction over the structure
depth d. From Eq. (2), Δneff x; y can be determined with
a knowledge of Δφx; y and d. The structure depth d can
be deduced through microscopy of the modified region,
while the induced phase change can be obtained using
the Fourier relation between the element plane and
the far-field distribution. Since the phase information
of the far field is usually not accessible, it is not possible
to reconstruct Δφx; y directly. In our approach an
iterative Fourier transform algorithm (IFTA) [12] is
applied, capable of retrieving the Δφx; y distribution
numerically from a measurement of the far-field intensity. In general, single FLDW features cause only weak
scattering and therefore marginal diffraction efficiency.
It is thus advantageous to fabricate multiple replica
separated with a known periodicity (px , py ). Because
of the Fourier relation, it leads to a discretely sampled
diffraction pattern in the far field. The intensity of the
corresponding diffraction orders can then easily be measured with a power meter (see Fig. 1). Eventually, the
implementation into the numerical algorithm yields the
refractive index distribution Δneff x; y averaged over
the individual replica. A uniform feature replication is
consequently essential in order to avoid artifacts in the
reconstruction. The lateral resolution of Δneff x; y is
limited by the number of measured orders N × M by
Δxmin ; Δymin ≃
2px ; py :
N − 1; M − 1
(3)
To validate this concept, we have first applied it to a
one-dimensional surface profile having a known refractive index distribution. The surface structure chosen consisted of a top hat profile repeated periodically to form a
binary grating. This had a period of 20 μm and a duty cycle of 0.55, emulating feature sizes similar to FLDW structures. These structures were fabricated by electron beam
lithography. A 950 PMMA A5 layer with a thickness of
approximately 500 nm and a refractive index of 1.49 at
543 nm wavelength was spun on a sapphire glass
substrate. The layer was patterned using a Leica EBPG
5000 electron beam. To obtain an accurate reference
of the refractive index distribution, the resulting surface
profile was measured using an atomic force microscope capable of resolving the grating height up to a
few nanometers. The surface profile, shown for one
grating line in Fig. 2 (black line), was characterized by
a grating height of 547 4 nm and a line width of
10.8 0.6 μm. Applying the Fourier method to this
sample, the grating was illuminated with a 543 nm He–Ne
laser. The resulting intensities in the different diffraction
orders were measured using a power meter. The measured distribution was inserted into the IFTA, producing
the reconstructed phase change shown in Fig. 2 (blue
dashed line). This phase distribution directly matches
the measured surface profile. The width of phase change
11.0 0.8 μm correlates well with the AFM data.
Furthermore, when the value of the phase change
Fig. 2. (Color online) 950 PMMA A5 surface profile for one
grating line measured with an AFM (solid curve) in comparison
with the reconstructed phase change (dashed curve) using the
Fourier reconstruction method.
3.11 0.15 rad, determined from Fig. 2, is inserted into
Eq. (2), and combined with the measured grating height,
it leads to a calculated change in refractive index of
0.49 0.03. This agrees exactly with the difference
in refractive index between that of air and of the PMMA
used to fabricate the surface grating structure.
In a second demonstration of this new approach to
measure Δn, FLDW structures in borosilicate glass
(Corning EAGLE 2000) were analyzed. The structures
comprised a parallel array of linear waveguides written
into a bulk borosilicate glass sample forming a planar,
one-dimensional grating. The grating lines separated with
a 20 μm period were written using an amplified 1045 nm
fiber laser (IMRA America μJewel D-400-VR) with pulse
energies of ∼1.5 μJ at a repetition rate of 200 kHz,
focused with a 0.4 NA objective, 225 μm below the glass
surface with a translation speed of 1 mm∕ s.
The dimensions of the refractive index changed region,
shown in Fig. 3, were measured with an optical transmission microscope. The cross section in Fig. 3(b) shows a
total depth d of 40 μm and a maximum width of 10 μm.
Additionally, smaller subfeatures within the modified
region, such as the 2 μm wide center area, can also be
observed. The varying brightness is caused by dissimilar
scattering and deflection in different parts of the modified area. This observation indicates an inhomogeneous
refractive index distribution within each waveguide [13].
The intensity of the far-field diffraction pattern was measured, where 43 separated orders could be distinguished.
Fig. 3. Microscopic image of femtosecond-laser induced
refractive index change. (a) Top view x; y of three grating
lines and (b) cross section x; y of single grating line.
July 15, 2012 / Vol. 37, No. 14 / OPTICS LETTERS
Fig. 4. (Color online) Refractive index change determination
of a repeated waveguide written in borosilicate glass with a
femtosecond-laser. (a) Far-field amplitude of resulting diffraction pattern using an He–Ne laser at 543 nm. (b) Determined
effective refractive index change distribution, which represents
an averaged induced waveguide.
The corresponding field amplitudes are shown in
Fig. 4(a). In addition to the diffraction efficiency of approximately 3%, a slight asymmetry in the distribution
was noticed. Since the grating was adjusted perpendicular to the incident laser beam, this indicated that the modified area itself was not perfectly symmetric. The phase
change was reconstructed using the IFTA, and by applying Eq. (2), the distribution of the induced effective was
determined. Here we assumed that an overall positive Δn
was produced, following Eaton et al. [13] for similar writing parameters. Figure 4(b) shows a well-separated peak
that corresponds to an average grating line with a main
feature size of 10 μm matching well the optical microscope measurements in Fig. 3. According to Eq. (3), a lateral resolution of ∼1.0 μm was obtained. A maximum
effective refractive index change of 2.2 0.1 × 10−3
was determined, which is of the same order as the results
presented in [14]. In addition, a reduced effective refractive index modification of 0.9 0.1 × 10−3 in the central
part of the modification was detected. Its width of 2 μm
agreed well with the width of the dark subarea in the center of the cross section in Fig. 3(b). This could be referred
to a region where a negative refractive index modification was induced. Measurements by Eaton et al. [14]
showed a similar phenomenon in borosilicate glass.
Finally as a direct comparison, the grating method
used by Mailis et al. [6] to estimate Δn was applied to
the structures used here. This approach gave a value
of ∼0.8 × 10−3 , which is close to the refractive index
change in the reduced central region but less than
that of the overall complex structure. The principal difference between the grating method and the Fourier
3005
reconstruction approach is the assumption of a simple
sinusoidal refractive index distribution. Consequently,
the grating method only provides a lateral average over
the whole structure. The Fourier method we describe
here is thus of added value in resolving fine differences
in the refractive index change.
In conclusion, we have demonstrated a practical
method of measuring femtosecond laser induced refractive index changes to an accuracy of 10−4 using an
iterative Fourier transform technique that renders an accurate estimation of the refractive index profile with high
spatial resolution across small micrometer-scale structures. No additional sample preparation is required. This
method should find application in the fabrication of many
two- and three-dimensional laser written refractive index
structures in transparent media, including the fabrication of efficient DOEs, fiber Bragg gratings, and other
photonic structures.
This work was performed as part of the Atlantis MILMI
program, funded jointly by the U.S. Department of Education and the European Commission, and was funded in
part by the National Science Foundation and the State of
Florida. Useful discussions with M. Ramme and Dr. I.
Mingareev are gratefully acknowledged.
References
1. S. Nolte, M. Will, J. Burghoff, and A. Tünnermann, J. Mod.
Opt. 51, 2533 (2004).
2. T. Fernandez, S. Eaton, G. Della Valle, R. Vazquez, M.
Irannejad, G. Jose, A. Jha, G. Cerullo, R. Osellame, and
P. Laporta, Opt. Express 18, 20289 (2010).
3. T. Gerke and R. Piestun, Nat. Photon. 4, 188 (2010).
4. J. Choi, M. Ramme, T. Anderson, and M. C. Richardson,
Proc. SPIE 7589, 75891A (2010).
5. I. Mingareev, R. Berlich, T. J. Eichelkraut, H. Herfurth,
S. Heinemann, and M. C. Richardson, Opt. Express 19,
11397 (2011).
6. S. Mailis, A. Anderson, S. Barrington, W. Brocklesby,
R. Greef, H. Rutt, R. Eason, N. Vainos, and C. Grivas,
Opt. Lett. 23, 1751 (1998).
7. D. Homoelle, S. Wielandy, A. Gaeta, N. Borrelli, and
C. Smith, Opt. Lett. 24, 1311 (1999).
8. C. Voigtländer, D. Richter, J. Thomas, A. Tünnermann, and
S. Nolte, Appl. Phys. A 102, 35 (2011).
9. I. Mansour and F. Caccavale, J. Lightwave Technol. 14, 423
(1996).
10. A. Zoubir, C. Lopez, M. Richardson, and K. Richardson, Opt.
Lett. 29, 1840 (2004).
11. E. Glytsis, J. Opt. Soc. Am. A 19, 702 (2002).
12. R. W. Gerchberg and W. O. Saxton, Optik 35, 237 (1972).
13. S. Eaton, M. Ng, J. Bonse, A. Mermillod-Blondin, H. Zhang,
A. Rosenfeld, and P. Herman, Appl. Opt. 47, 2098 (2008).
14. S. Eaton, H. Zhang, M. Ng, J. Li, W. Chen, S. Ho, and P.
Herman, Opt. Express 16, 9443 (2008).